Download Auto-connection technique in primitive part modelling

 S1 
0
0
0  
 C( 1,1 ) C( 1,2 )
 D( 1 ) 
S 2  
C( 2 ,1 ) C ( 2 ,2 ) C( 2 ,3 )


0
0
D( 2 )

 × WM  = 

 0
C( 3 ,2 ) C ( 3 ,3 ) C( 3 ,4 ) C ( 3 ,5 ) 
  D( 3 )

 W1  

0
C( 4 ,3 ) C( 4 ,4 )
0  
D( 4 )
 0
W 0  
(8)
This overall matrix template for the pipe can be derived from the conventional approach by
setting the conservation equations. At the same time, the coefficients of this matrix equation
can be given by the summation of primitive part coefficients. Figure 4 shows an example of the
data flow process. In this example, the user made connections between nodes of different
PP’s in the order of PP4.3, then PP1.1, and then PP2.4. Accordingly, the node order (or the
row number of the matrix) from 1 to 5 is for S2, WM, W1, W0, and S1. This is purposefully
made different from those in equation (8) to demonstrate that the data structure can support
different assigned node order and can arrive at the same required values of the overall matrix
coefficients.
If a specific simulation task is to investigate the thermal behavior of the insulated water pipe
with respect to changes in inlet water temperature conditions, equation (8) or the equivalent
one built by PP’s becomes the global matrix of the defined problem. The program identifies the
coefficients index of auto-connected PP and dynamically set the matrix coefficients in the
course of simulation. Figure 5 gives the simulation results of two specific flow cases:
o
o
Case1: Step change in inlet water temperature from 5 C to 10 C.
o
o
Case2: Linear change in inlet water temperature from 5 C to 10 C in 10 seconds.
These results are exactly the same as the one developed directly from the conventional
component-based modeling approach.
4. Flexibility Vs Complexity
The benefit of applying PP’s is to allow a flexible and generalized representation of the HVAC
system under study. A pre-requisite for the convenient use of the PP’s is then to make
available in the input interface an adequate data set of commonly used parameters in system
simulation. These will include technical data such as physical dimensions (like length,
thickness, volume etc.), material properties (like density, specific heat, conductivity etc.) and
heat and mass transfer coefficients at various flow conditions (like piped flow, wind induced
airflow, natural convection, wet surfaces etc.). The input parameters should be readily
availabl e at a friendly and practical user interface. For those less commonly used parameters,
the user is allowed to insert the additional information based on their own technical knowledge
or experimental research of the situation under study.
9